Does

kotlin.mpp.enableGranularSourceSetsMetadata=true

work with Android Studio? I tried creating a common sourceset that was common between macosx64 and iosX64 and iosArm64, and I couldn’t get any code completion. Is this expected to work?

I'm wondering if I am doing things correctly. Below is what I have done so far (the IDE doesn't show any errors):

public actual fun saveToBuffer(
        type: String,
        options: Map<String, String>,
        bufferSize: Int,
        error: Error?
    ): Pair<ByteArray, Boolean> = memScoped {
        val tmpBuffer = allocArray<CPointerVar<ByteVar>>(bufferSize)
        val arraySize = options.keys.size
        val keys = allocArray<CPointerVar<ByteVar>>(arraySize)
        val values = allocArray<CPointerVar<ByteVar>>(arraySize)
        options.keys.forEachIndexed { pos, item -> keys[pos]?.pointed?.value = item.toByte() }
        options.values.forEachIndexed { pos, item -> values[pos]?.pointed?.value = item.toByte() }
        val errorSet = gdk_pixbuf_save_to_bufferv(
            pixbuf = gdkPixbufPtr,
            type = type,
            error = cValuesOf(error?.gErrorPtr),
            option_keys = keys,
            option_values = values,
            buffer_size = null,
            buffer = tmpBuffer
        ) == TRUE
        val buffer = ByteArray(bufferSize)
        @Suppress("ReplaceRangeToWithUntil")
        (0..(bufferSize - 1)).forEach { pos ->
            val item = tmpBuffer[pos]
            if (item != null) buffer[pos] = item.pointed.value
        }
        buffer to errorSet
    }

1. Are manually allocated C arrays in Kotlin Native null terminated? 2. Is buffer a shallow copy of tmpBuffer? 3. Is there an easier way to do a shallow copy of a byte buffer that has been manually allocated to a new copy of ByteArray?

Trying out some XCB in kotlin with cinterop, and i am struggling to set a window atom.

xcb_change_property(
    this.connection,
    XCB_PROP_MODE_REPLACE.toUByte(),
    window,
    windowTypeReply.pointed.atom,
    XCB_ATOM_ATOM,
    32,
    2,
    windowTypeDockReply.pointed.atom // THIS needs to be a CValuesRef
)

The

windowTypeDockReply

is of the type

CPointer<xcb_intern_atom_reply_t>

, which has a property

atom

of the type

xcb_atom_t

which can be accessed through the pointed property. The function

xcb_change_property

in c needs a

xcb_atom_t

as its last parameter. In Kotlin however it requires a

kotlinx.cinterop.CValuesRef<*>?

. So i am wondering how i can make

windowTypeDockReply.pointed.atom

into a

CValuesRef

?

Is there something like Touchlab's Stately Isolate that wouldn't block the calling thread? Possibly exposing a suspend interface?

Does anyone know why

mutableListOf

would not be available in the

commonMain

namespace? This is incredibly frustrating, since i think this worked earlier, and now it doesn't. Invalidated caches, restarted IDE and computer, nothing seems to help 😕

I think the latest IDE/Plugin builds are broken. But what's shocking me the most is that nobody talks about it, as if nobody is actually using it...

https://github.com/Doomsdayrs/kotlinx-gtk/releases/tag/0.1.0-alpha I feel like the library has matured enough to be considered plausible to create an application with it. While some features might be needing, with time and more people helping I feel like this will be the definitive gtk / kotlin library 😄

Any advice on how to speed up kotlin multiplatform builds? When building the first time via CLI, my package takes 10+ minutes, and subsequent builds take over 5 minutes. My package only has a few small classes.

hi, what is the state of

kotin-coroutines-native-mt

? I’m trying to update to 1.5.0 from 1.4.2, and my tests are failing left and right with

mutation attempt of frozen kotlinx.coroutines.flow.StateFlowImpl

Hi I have two project "library" and "application" "library" contants some class and some method in this class. In "application" I trying to call the method and getting error like

Compilation failed: Symbol for public <http://pw.binom.io|pw.binom.io>.httpClient/HttpClient.request|-4729544346512717501[0] is unbound

Then I renamed

HttpClient.request

to

HttpClient.connect

and build it. Then I again try to compile "application" and get error: method

request

not found. It is fine, because I renamed method. I fix call

request

to

connect

in "application" and tryed to build "application" again. And in this turn I got same error:

Compilation failed: Symbol for public <http://pw.binom.io|pw.binom.io>.httpClient/HttpClient.request|-4729544346512717501[0] is unbound

Maybe I must clear some cache? I made

gradle clean

on both project. It not work.

I made a tutorial video to show off how easy it is to make a GTK application with Kotlin/Native using kotlinx-gtk What do you all think?

Made a minor change to

kotlin-native

but does not seem to take effect. I think it might be helpful to click the

load gradle changes

button, but an error was reported, what should I do?

Hello, I am successfully using kotlin native since months for mobile cross platform development and I have an ios background. There is still something I don’t understand…How to know what’s the version of iOS APIs included in a specific Kotlin version? For example I see that when new version of iOS will come out, kotlin will not include immediately brand new APIs. Is that correct? Or Kotlin just translate current ios sdk installed on your machine into klibs so kotlin native APIs are always up-to-date?

We have a iOS library written in Kotlin and we have a few customers that uses it on production. On of our client is getting crash reports from users related to our library (crash report and integration code in thread). From the code and crash reports it is very hard to tell what could go wrong - the crash happens in during the initialization process of class that keeps reference to the our library’s main class. The first line from the crash report is:

0  AudioburstMobileLibrary    	0x1035f2f0c kfun:kotlinx.coroutines#<get-mainThread>(){}kotlinx.coroutines.MainThread + 1289996 (Thread.kt:31)

However I don’t have a crash message, so I can’t be sure what might go wrong there. Is there any way to investigate this problem? Did you ever see something similar to this? The crashes are not reproducible by the client, but it affects like 70 users

Any thoughts on supporting Swift's new async/await to map to coroutines?

Encountered an issue where the cinterop tool isn't generating the bindings for the GtkPrinter API ( https://developer.gnome.org/gtk3/3.20/GtkPrinter ). Some of the other GTK printing type APIs (eg GtkPrintSettings - https://developer.gnome.org/gtk3/3.20/GtkPrintSettings) have generated bindings. Does anyone know of any cases where the cinterop tool fails to generate bindings for something that isn't a non trivial C Macro?

Has anyone tried to get https://www.keystone-engine.org working in kotlin-native and mingwX64? I have a particularly weird result here that I can't wrap my head around. Given this c code https://www.keystone-engine.org/docs/tutorial.html I have translated it to https://github.com/Syrou/kotlin-native-keystone/blob/main/src/nativeMain/kotlin/main.kt which compiles and runs, problem is, the result is wrong. The size, and the statement counter seems to be correct. Everything seems to setup properly, yet again, no matter what I do the result is wrong.

INC ecx; DEC edx = 41 4a

is the expected outcome. However it always returns

INC ecx; DEC edx = 60 43

Does anyone have an idea to what is going on? Have I done any obvious misstakes or is the generated bindings somehow messing up?

Obviously probably not possible to answer this exactly at this point but roughly when should we hope to see direct Swift interop.....I know it's in postponed column still in https://kotlinlang.org/docs/roadmap.html#roadmap-details but seems to be feature many people are clamouring for

I'm trying to call a Objective-C method that takes a NSError. All the samples I see show,

memScoped {
    val errorRef = alloc<ObjCObjectVar<NSError?>>()
    val result = function(errorRef.ptr)
}

But I get error message saying

None of the following functions can be called with the arguments supplied: 
public open fun alloc(size: Int, align: Int): NativePointed defined in kotlinx.cinterop.MemScope
public final fun alloc(size: Long, align: Int): NativePointed defined in kotlinx.cinterop.MemScope

Trying to see how Swift's new async/await could be hooked in to shared KMM code and noticed by chance that following (in thread) works was was sort of curious why 🙂 ...

Up until recently, when my ios app crashed with some InvalidMutabilityException, I could read the kotlin error from the xcode console. Right now it doesn't say anything more than

Function doesn't have or inherit @Throws annotation and thus exception isn't propagated from Kotlin to Objective-C/Swift as NSError.

And the stacktrace is obviously rubbish. I cannot pinpoint when it exactly changed, any ideas what might be causing it?

I got a little bit tired of banging my head against C++ and decided to at least try to do a kotlin wrapper. The library I am using is in C++, but seems to have plain c interfaces. I managed to create interop file, but the header depends on C++ stdlib files like

cmath

,

memeory

etc. How can I link them as well?

No external dsym's are created (or needed).

also, has anyone gotten cocoapods to work? Trying to add Firebase to my shared module, and was only able to get it to work if i made the framework static. If the framework is dynamic, I get this error:

> Task :shared:linkDebugFrameworkIos FAILED
e: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/ld invocation reported errors
Please try to disable compiler caches and rerun the build. To disable compiler caches, add the following line to the gradle.properties file in the project's root directory:

    kotlin.native.cacheKind=none

Also, consider filing an issue with full Gradle log here: <https://kotl.in/issue>
The /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/ld command returned non-zero exit code: 1.
output:
Undefined symbols for architecture x86_64:
  "_OBJC_CLASS_$_FIRDatabase", referenced from:
      objc-class-ref in result.o
  "_OBJC_CLASS_$_FIRApp", referenced from:
      objc-class-ref in result.o
ld: symbol(s) not found for architecture x86_64

FAILURE: Build failed with an exception.

@Artyom Degtyarev [JB] - With the new linker it doesn't like dealing with symbolic links (often used in linking with the Linux ARM targets and the linuxX64 target). Is there a way to get the new linker to recognise a symbolic link to a Shared Library? Do note that the linking issues don't occur when Kotlin is downgraded from 1.5.10 to 1.4.31 (uses the old linker).

Is there going to be an in depth guide on linking with Kotlin Native (on the Kotlin website)?

I believe I found a kotlin-native bug. I want to bounce it off of people here before I report it, to see if I'm just doing something wrong or perhaps expecting too much. When I build this code sample (let me know if anyone wants the entire reproducing project) with kotlin native, at least on linux_x64, and then run the resulting binary, it will print out bb_b and bb_c just fine, but bb_a (which comes before bb_b alphabetically!) will then cause a segfault, because bb_a.number is uninitialized. I'm using a trivial wrapper class because the issue doesn't seem to manifest with types that can become C types. Of course, the issue also doesn't manifest when running on jvm.

fun main(args: Array<String>) {
    NumbersByBit.values().forEach {
        println("The number ${it.name} is: ${it.number.num}")
    }
}

enum class NumbersByBit(val number: NumWrapper) {
    bb_b(NumWrapper(6)),
    bb_c(bb_b.number),
    bb_a(bb_b.number),
}

class NumWrapper(val num: Int) {
    operator fun plus(other: NumWrapper) = NumWrapper(num + other.num)
}

So is this a bug? Or are self-referential enums out-of-scope for kotlin native? Update: Based on above messages, I tried kotlin 1.4.31 and the same error occurs. When I test the binaries with valgrind, I see an invalid read:

The number bb_c is: 6 
==895440== Invalid read of size 4 
==895440==   at 0x443ED1: Init_and_run_start (in /home/alex/IdeaProjects/NativeError/nativeError.kexe_1.5.10~) 
==895440==   by 0x4A300B2: (below main) (libc-start.c:308) 
==895440== Address 0x8 is not stack'd, malloc'd or (recently) free'd

The invalid read error looks the same when running the kotlin 1.4.31 version.

Commonizer for linux seems to have broken in 1.5.10. I get the same errors on my local WSL (ubuntu) as well as on GH Actions. Has anyone else experienced something similar? Is there a bug for this already (so I wouldn't raise a duplicate)

i am trying to set the modification times on a file apparently in win32 api there exists

dwHighDateTime

and

dwLowDateTime

which are there so that a 64 bit value can be split into the lower and upper 32 bit how would i split a kotlin Long into 2 UInt (DWORD) values ? or can i just write a Long value to the

FILETIME

struct via some unsafe operation ?

How can I create wrapper classes for generated c structs? Whats the common pattern? I want to hide away the usage of pointers, c values, and such and leave that as internals. Trying to create a kotlin port of a c lib which I then want to call within a kotlin jvm project. Not sure if this is possible or proper way of doing so though as I have little experience.